Yandex-8B-WMT26

Yandex-8B is our constrained submission to the WMT26 General Machine Translation task. It translates English into Russian, Belarusian, Kazakh, and Armenian. The model is an approximately 8B-parameter causal language model fine-tuned from YandexGPT-5-Lite-8B-pretrain.

The released checkpoint is the supervised instruction-tuned model submitted to WMT26. It was trained on approximately one million direct and instruction-conditioned translation examples across the four directions. The targets were generated by Yandex, our 235B unconstrained WMT26 submission. The mixture also included dedicated JSON and HTML examples. We also tested Contrastive Preference Optimization. We did not use a CPO checkpoint for the submission because it did not provide a consistent improvement.

Model details

Property	Value
Model type	Decoder-only causal language model
Parameters	Approximately 8B
Base model	yandex/YandexGPT-5-Lite-8B-pretrain
Source language	English
Target languages	Russian, Belarusian, Kazakh, Armenian
Context length	32,768 tokens
License	YandexGPT-5-Lite-8B License

Usage

The repository includes inference.py, which supports plain translation prompts, the four WMT26 domain prompts, and custom instructions. It uses greedy decoding and prints only the generated translation to standard output.

Install the required packages:

python -m pip install "torch>=2.1" "transformers>=4.46.3,<5" accelerate sentencepiece packaging

Translate a string into Russian:

python inference.py \
  --target ru \
  --text "The agreement will enter into force next month."

Use one of the WMT26 domain instructions:

python inference.py \
  --target kk \
  --domain news \
  --text "The committee announced the results on Tuesday."

The supported domain values are social, speech, news, and software. The source text can also be supplied through standard input. Use --prompt or --prompt-file to provide a custom instruction.

Transformers example

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "yandex/Yandex-8B-WMT26"
tokenizer = AutoTokenizer.from_pretrained(model_id, use_fast=False)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype=torch.float16,
    device_map="auto",
).eval()

prompt = """Переведи с английского на казахский:
The committee announced the results on Tuesday."""
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
output = model.generate(**inputs, max_new_tokens=512, do_sample=False)
print(tokenizer.decode(
    output[0, inputs["input_ids"].shape[1]:],
    skip_special_tokens=True,
))

Evaluation

We evaluate the released checkpoint against five public models with at most 20B parameters. These are local evaluation results, not official WMT26 scores.

BOUQuET

The following paragraph-level results use the BOUQuET references. MetricX uses the google/metricx-24-hybrid-xxl-v2p6 checkpoint.

ChrF++ (higher is better)

System	en-ru	en-be	en-kk	en-hy
Yandex-8B	62.3	58.0	55.3	53.5
TranslateGemma-12B	62.2	54.9	47.9	50.1
Qwen-3.5-9B	60.1	49.1	48.3	49.6
MADLAD-400-10B	59.4	54.2	46.8	50.3
NLLB-200-3.3B	57.4	52.1	47.8	52.8
GPT-OSS-20B	51.9	29.3	49.0	12.9

MetricX-24-XXL (lower is better)

System	en-ru	en-be	en-kk	en-hy
Yandex-8B	1.58	3.87	3.44	5.41
TranslateGemma-12B	1.52	3.83	5.60	6.30
Qwen-3.5-9B	2.22	5.73	5.52	7.32
MADLAD-400-10B	4.79	5.56	5.86	7.13
NLLB-200-3.3B	6.01	7.63	7.10	6.35
GPT-OSS-20B	2.72	6.16	5.82	9.18

WMT25 General MT source paragraphs

We also translate the official English source paragraphs from the WMT25 General MT task and evaluate them with ORBIT-SC using GPT-5.4 as a single judge. The MQM score is computed as 5 × Major + Minor from the predicted error spans.

Accuracy (higher is better)

System	en-ru	en-be	en-kk	en-hy
Yandex-8B	81.6	69.1	69.9	52.2
TranslateGemma-12B	73.9	59.1	47.2	41.8
Qwen-3.5-9B	69.6	51.1	46.2	39.1
MADLAD-400-10B	32.5	40.5	40.8	28.3
NLLB-200-3.3B	39.7	33.8	32.2	35.0
GPT-OSS-20B	62.4	40.9	40.8	30.2

Fluency (higher is better)

System	en-ru	en-be	en-kk	en-hy
Yandex-8B	85.3	69.5	72.0	54.0
TranslateGemma-12B	81.2	64.4	52.5	48.1
Qwen-3.5-9B	72.7	51.8	51.4	44.8
MADLAD-400-10B	32.5	44.8	48.4	37.5
NLLB-200-3.3B	40.3	33.8	35.7	36.4
GPT-OSS-20B	63.5	39.3	42.3	29.4

MQM (lower is better)

System	en-ru	en-be	en-kk	en-hy
Yandex-8B	11.3	24.2	22.3	37.6
TranslateGemma-12B	17.9	30.8	40.9	43.8
Qwen-3.5-9B	21.4	38.5	40.8	46.4
MADLAD-400-10B	34.5	38.6	40.1	31.7*
NLLB-200-3.3B	39.5	46.0	45.5	46.2
GPT-OSS-20B	27.2	46.8	44.5	52.4

* The MADLAD English-to-Armenian MQM value is affected by the count-based aggregation of a small number of long critical spans. Its low value should not be interpreted as strong translation quality.

License

The model is distributed under the YandexGPT-5-Lite-8B License. Review the license before using or redistributing the model.